Question

So here is the basic code:

    namespace WindowsFormsApplication1
{
    public partial class Form1 : Form
    {
        protected Random random;

        public Form1()
        {
            InitializeComponent();
            random = new Random();
        }

        private void Form1_Load(object sender, EventArgs e)
        { }

        private void button1_Click(object sender, EventArgs e)
        {
            bool button1Clicked = true;
            if (button1Clicked == true) { ITIpanel.Visible = true; }
        }

        private void ITIpanel_Paint(object sender, PaintEventArgs e)
        {
            ITItimer.Enabled = true;
        }

        private void ITItimer_Tick(object sender, EventArgs e)
        {
            double rand = random.NextDouble();
            if (rand < .50d) { bluestimPanel.Visible = true; }
            else if (rand > .5d) { redstimPanel.Visible = true; }

            ITItimer.Enabled = false;
        }

        private void bluestimPanel_Paint(object sender, PaintEventArgs e)
        {
            Trialtimer.Enabled = true;
        }

        private void redstimPanel_Paint(object sender, PaintEventArgs e)
        {
            Trialtimer.Enabled = true;
        }

        private void Trialtimer_Tick(object sender, EventArgs e)
        {
            bluestimPanel.Visible = false;
            redstimPanel.Visible = false;
            Trialtimer.Enabled = false;
            ITIpanel.Visible = true;
        }
    }
}

As you can see, the program itself is fairly straight forward. At the tick of the ITItimer the red or blue panels occurs at random. I want to modify this such that if the ITItimer ticks a total of 10 times, the red and blue panels will both have occurred 5 times each.

I have been researching this for a week or so and have yet to find a solution. Any ideas on how I could best accomplish this?

I actually got the following to work:

double rand = random.NextDouble();
if (rand < .50d && blue < 5) { bluestimPanel.Visible = true; }
else if (blue == 5) { redstimPanel.Visible = true; }
if (rand > .5d && red < 5) { redstimPanel.Visible = true; }
else if (red == 5) { bluestimPanel.Visible = true; }
if (red >= 5 && blue >= 5) { panel1.Visible = true; } 

It isn’t exactly the prettiest thing in the world. But it gets the job done.

Answer 1

Random numbers using most normal library routines are a low-quality source of pseudorandomness. If this is for a randomized scientific study, this will be a flaw in your protocol design.

The approach that I would recommend would be to consider this a method of randomly arranging a session of at least N trials, where there are X trial types.

The below is pseudocode for illustration of the concept.

Let MinimumTrials be N MOD X + X
Let SessionList be a List<Trial>
For Each TrialType 
    add X instances of that trial type to SessionList
Shuffle(SessionList)

Then your session engine can call the individual Trials as it walks through the SessionList to have an even distribution of possible trial orders. Note that Shuffle is an operation which requires a certain degree of finesse to get right, searching on SO is a good starting point for that.